COMMUNICATIONS T O T H E EDITOR A NEW CLASS OF POTENT CENTRAL NERVOUS SYSTEM DEPRESSANTS
chloroethyl isocyanate7and a-naphthylamine or by treating 1-( 1-naphthyl) -3-(2-hydroxyethyl) -urea with thionyl chloride in refluxing chloroform. Sir: Acidic degradation of I with boiling 3N hydrochloric Slthough the first 2-amino-2-oxazolines (pseudo- acid leads to V. ureas) were synthesized over fifty years ago, The iodo analog, 1-(l-naphthyl)-3-(2-iodoethyl)compounds containing that grouping have until grea (VI), m.p. 176.0-178.O0; , : :A: 3.09, 6.17, 6.39, now received only cursory pharmacological exami- 6.60, 6.68; (Anal. Calcd. for C ~ ~ H I ~ X Z C, O I 45.9; : n a t i ~ n . We ~ now wish to report the synthesis, in HI 3.85; N, 8.24. Found: C, 45.9; 13, 3.9-1-; the course of a systematic search for new psycho- N, 8.25) was prepared by refluxing I1 with exceqs therapeutic agents, of the previously unknown sodium iodide in acetone s o l ~ t i o n . ~ 2-(l-naphthylamino)-2-oxazoline (I). This comFollowing oral or parenteral administration OF the pound represents a new structural type of central oxazoline I to cats, dogs and monkeys, markctl nervous system depressant, which displays a unique quieting was observed accompanied by mtisclc. pharmacological profile. relaxation and increased ease of handling. In Treatment of a-naphthyl isocyanate with p- potentiation of hydroxydione (21-hydroxypregbr~moethylamine~ in benzene solution led to 1- nanedione sodium succinate) anesthesia, I proved ( 1-naphthyl) -3-(2-bromoethyl) -urea (11), m.p . somewhat more potent than either reserpine or 146.8-147.2’; . : : A: 3.10, 6.19, 6.41, 6.60, 6.70; chlorpromazine, while all three drugs depressed thc> (-4nal. Calcd. for C I ~ H I ~ N Z OC, B ~53.3; : HI 4.47; spontaneous motility of mice to a similar degree. N, 9.56; Br, 27.3. Found: C, 53.1; HI 4.37; Spinal cord depressant activity could be demonN, 9.76; Br, 27.1). Alternatively, I1 could be strated with lower doses of I than wcre recluiretl synthesized by treating P-bromopropionyl chloride5 with the distant structural relative, 2--a1nitio-~?with sodium azide under thermal-rearrangement chlorobenzoxazole.10 Metrazole convulsions werc conditions to afford the previously unknown P- facilitated by pretreatment with I, as is thc C ~ W bromoethyl isocyanate, which was not isolated but with reserpine.l’ allowed to react directly with a-naphthylamine ( 7 ) H . Wenker, THISJOURNAL, 68, 21308 (1936); W Siefken. i i i r r t , affording the desired compound. The urea I1 562, 7 5 (1949). (8) R.W. Charlton and A. R. D a y , J. Org. Chem., 1, 552 (193(?). readily underwent intramolecular cyclization in (9) I n animals the 8-bromoethyl urea I1 displays n spectrum of boiling waterE to give a solution of Z-(l-naphthyl- pharmacological actions similar to I, although much higher doses arc amino)-2-oxazoline hydrobromide, from which required. Since it can be demonstrated t h a t any efficient solvdytic dilute ammonium hydroxide precipitated the free medium (e.&, methanol, dimethylformamide) will effect t i l L ’ i l i w base I, m.p. 123.4-124.6; E:. 3.36, 3.55, 5.98, cyclization of the urea I1 t o t h e oxazoline I a t room temperature. i t O : highly probable t h a t a similar in vi110conversion is involved i t 1 6.08, 6.38, 6.64; (Anal. Calcd. for C I ~ H ] Z ~ \ ; ~tseems h e pharmacodynamics of 11. Further investigation of this n o d C, 73.6; H, 5.70; N, 13.2. Found: C, 73.5; metabolic transformation is contemplated. Neither the iodo nor HI 5.60; N, 13.1). chloro analog is as active as I1 when evaluated by the technique Treatment of I with anhydrous hydrochloric anesthesia potentiation in rats. T h e whole picture of ;)I B i U O tttinformation t o a n active species by intramolecular alkylation, as well : I acid in methylene chloride solution afforded the varying potency among halogen analogs is reminiscent of the 8-halo corresponding hydrochloride salt, m.p. 150.2- ethylamine adrenergic blocking agents (G. E. Ullyot and J. I.‘ k’erwirl 150.5O; . : : A: 3.25, 3.53, 6.00, 6.18, 6.2S, 6.33, “Medicinal Chemistry,” John Wiley and Sons, Inc., New York. S .1- , (3.41, which reverted in boiling 2-propanol to 1-(1- Vol. 11. 1956. p , 234). (101 I\’. H . Funderburk and R. T. Woodcock. Fed PYoc , 1 4 , 311 naphthyl)-3-(2-chloroethyl) -urea (IV), m.p. 145.6- (1955); K. Kamijo and G. B. Koelle, i b d , 14, 3Sii (1%:); I ) . 1’ 146.4; , ::A: 3.07, 6.14, 6.40, 6.65; (Anal. Calcd. Marsh, ibid.,14, 366 (1955). ( I 1) T h e detailed pharmacology of I will be published elsewhere for CI3Hl3iV20C1: C, 62.8; H, 5.27; N, 11.3. BARRY If.BLOOM Found: C, 63.1; HI 5.26; N, 11.1). The synLABORATORIES JOSEPH F. GARDOCK1 thesis of I V could also be accomplished from p- RESEARCH CHAS.PFIZIIRA N D Co., INC. DUNCAN E. IlmCimox (1) R . H. Wiley and L. L. Bennett, Jr., Chem. Reus., 44, 404 (1949); J. W. Cornforth. “Heterocyclic Chemistry,” 1‘01. 5, John Wiley and 1957, Vol. 11, pp. 384-388,390. Sons, Inc., New York, N. Y., (2) For more recent chemical studies. see A. F. McKay and coworkers, THISJOURNAL,78, 486, 1618, 6144 (1956); C a n . .I. Chena., 35, 8 (1957), and earlier papers. (3) C. J. Rose, H. A. Shonle and K. R. Chen, Phurmacrulicnl Archiues, 11, 81 (1940); R. Gebauer, German Patent 694,133 (July 26, 1940); C . A . , 38,5259 (1941). (4) hl. Engelmann, U. S. Patent 2,027,031 (Jan. 7, 1936); C . A . , 30, 1519 (1936). (5) C. S . Hamilton and C. L. Simpson, THISJ O U R N A L , 51, 3158 (19’29); R. Dahlbom, Acta Chem. Scand., 7, 873 (1953). (6) In accord with earlier observations in t h e 8-haloalkylurea series,
RROOKLPX.NEWYORK GERALD n.~ , A I : I I A C l I RECEIVED JUNE 17, 1957
BIS-( CY CLOPENTAD1ENYL)-TITANIUMDICHLORIDE -ALKYLALUMINUM COMPLEXhS AS CATALYSTS FOR THE POLYMERIZATION OF ETHYLENE Sir :
The recent publications of Katta and coworkers’,? on the use of bis-(cyclopentadien4.1)-,
(1) G . N a t t a . P . Pino, G. Mazzanti, U. Giannini, E, I l a n t i c a nnd alkali-catalyzed cyclization of I1 proceeded via alkylation on nitrogen 31. Peraldo, C h i m e . rizd. ( M i l a n ) , 39, 19 (1957); C . A . , 51, 7010 t o produce l-(l-naphthyl)-2-imidazolidone(III), m.p. 180.4-181.6°, (1957). XKBr ms.. 3.17, 3 . 3 2 , 3.57, 5 . 9 7 , 6.31, 6 . 7 2 (Anal. Calcd. for CIIHIZNZO: (2) G . N a t t a , P. Pino, C . hlazzilnti a n d U. Giannini, ‘l‘rlrs TOIIIWAI.. C , 7 3 . 6 ; H.5 . 7 0 ; N , 13.2. Found: C 74.2; H , 5 . 6 1 ; N, 13.3). 79, 2975 (1957).
Sept. 20, 1057
COMMUNICATIONS TO THE EDITOR
titanium derivatives as catalysts for the low-pressure polymerization of ethylene prompt US to report some related work which we carried out several years ago. I n several respects our results differ considerably from theirs. An orange solution of bis-(cyclopentadieny1)titanium dichloride in toluene (5 millimoles per liter) was treated with two moles of diethylaluminum chloride. The color immediately changed to a dark red, followed by a gradual change over a period of about an hour a t room temperature to a green and finally to a blue solution. During this time ethane was evolved. (With triethylaluminum the formation of the blue solution was practically instantaneous ; the slow reaction reported by Natta can be attributed to the insolubility of the dichloride in heptane.) Replacement of the toluene by n-heptane yielded a clear blue solution, from which a blue crystalline solid could be isolated by chilling in Dry Ice. After several recrystallizations it melted a t 80-go’, apparently with some decomposition. Although the compound seemed to be thermally stable a t room temperature, it was extremely sensitive to traces of oxygen. The compound is a complex of (C5H5)2Ti(III)Clwith what appears to be aluminum sesquichloride. A nul. Calcd. for (C5H5)2TiCl.1/2(C2H5) 2AIC1-1’2C2H5AlCl,: C, 46.28; H I 5.23. Found: C, 46.36; H , 5.26. Analysis of a heptane solution gave the ratios (normalized for total titanium) Ti(II1 or less), 0.95; Al, 1.05; C1, 2.63. Magnetic susceptibility measurements showed the presence of one unpaired electron, while hydrolysis with thoroughly deaerated dilute mineral acid gave a green aqueous solution; the (C6H5)*Ti(I1I)ion is reported to be green3 There seems to be little doubt that, by analogy, the compound isolated by us and by Natta and co-workers from bis-(cyclopentadieny1)titanium dichloride and triethylaluminum has the structure (C5H5),TiCI.(C2H5),A1C1,the “sandwich” compound reacting with alkylaluminum compounds in a manner similar to titanium tetrachloride. The blue complex prepared with diethylaluminum chloride is a very poor catalyst for the polymerization of ethylene, in agreement with the observations of Katta and co-workers on the complex prepared with triethylaluminum. A fresh mixture of bis-(cyclopentadieny1)-titanium dichloride and diethylaluminum chloride, however, is a highly active catalyst, as is the blue complex if the ethylene contains a trace of oxygen4 The color changes in the latter case indicate quite definitely that the oxygen is functioning to form a tetravalent titanium compound. Thus, ethylene containing 0.003 mole yo oxygen was passed into a solution of 5 millimoles of bis-(cyclopentadienyl) titanium dichloride and 10 millimoles of diethylaluminum chloride in a liter of toluene a t 15-20’. The blue solution turned green, and 13 g. of polyethylene was formed in one hour. Under the same conditions ethylene containing 0.02.570 oxygen gave a brown solution; the addition of small amounts of a dilute solution of diethylaluminum chloride to maintain this color during the poly (3) G.Wilkinson and J M. Birmingham, THISJ O U R N A L , 1 6 , 4281 (1954). (4) D.S.Breslow, Belgian P a t e n t 551,283 (1957).
5073
merization resulted in the formation of 174 g. of polyethylene in one hour. Thus, these catalysts are fully as active as the usual Ziegler type. There seems to be little doubt from these results that this soluble catalyst system depends for its catalytic activity on the presence of a t least some tetravalent titanium. The polymers differ from polyethylene prepared with the usual Ziegler-type catalysts in being more linear (methyl content about 0.05% us. about 0.9%) and higher melting (137’ vs. 132’). DAVIDS. BRESLOW RESEARCH CENTER HERCULES POWDER COMPAXY WILMINCTON 99, DELAWARE ~YORMAS R. NEWBURG RECEIVED JULY 9, 1967
INHIBITION BY HYDROGEN PEROXIDE OF THE SECOND EXPLOSION LIMIT IN HYDROGEN-OXYGEN MIXTURES
Sir: During a recent investigation of the gas-phase decomposition of hydrogen peroxide’ in which we confirmed the partially homogeneous character of the reaction above 400°, we tried adding a number of gases to study their effect on the rate of peroxide decomposition. Of the various gases tested (H?, 0 2 ,NO, etc.) only hydrogen showed significant effects: a marked acceleration of the reaction in the temperature range 440-470’ a t total pressures in excess of 20 mm., in agreement with the findings of McLane2 in a flow system. At lower pressures we observed explosive reactions of the hydrogen with the oxygen from decomposition of the peroxide. However, a remarkable feature of these explosions was that they occurred only after all, or nearly all, the hydrogen peroxide had disappeared although the mixtures were within the explosion range3 for some time yet. This pointed out to a strong inhibiting action of hydrogen peroxide apparently not reported before. To check this point further we added some hydrogen peroxide vapor to stoichiometric hydrogen-oxygen mixtures and determined the second limit by the usual withdrawal method. The preliminary results confirm, indeed, that hydrogen peroxide is an efficient explosion inhibitor, roughly ten times as efficient as water vapor under the same conditions. For instance a t 458’ the second limit was lowered from 28 mm. to 17 mm. by about 0.01 mole per cent. of hydrogen peroxide, while a t 0.04 mole per cent., the explosions were suppressed entirely. There is some uncertainty in our results as to the exact concentration of peroxide a t the moment of explosion because its decomposition rate depends on such continuously changing variables as total pressure and concentration of hydrogen gas in the system. N‘ater vapor from the decomposing peroxide also acts as an inhibitor, but this is of minor importance in our case. That the above phenomenon has not been noticed(previous1y is no doubt due to the fact that, unless special care is exercised, hydrogen peroxide will decompose quickly through a heterogeneous mechanism before the necessary (1) P. A. Giguere and I. D. Liu, Can. J . Chcm., 36, 283 (1957). (2) C.K.McLane, J . Chem. Phys., 18,972 (1950). (3) B. Lewis and G. von Elbe, “Combustion, Flames and Explosions,” Academic Press, Inc., New York, N. Y., 1951.
